Systematic reviews of interventions for children with primary speech and language impairments (PSLI) have shown some evidence of effectiveness for some interventions. However, studies have identified a gap between the therapy speech and language therapists deliver and the evidence to support this current practice. Furthermore, for interventions to be appropriately targeted to subgroups of children, it is important to determine how interventions should be stratified for particular subgroups.
The aim of this systematic review was to identify which interventions are effective for which preschool children in which contexts, using a practice-based typology of interventions to group the studies.
Sixteen databases including Cinahl, Embase and Medline were searched for peer-reviewed English-language publications between January 1980 and November 2011. Search strings included key words pertaining to; speech, language, communication, developmental disorders, paediatric, and research design.
Fifty five thousand two hundred and seventy one publications were identified and reviewed and excluded, following Cochrane guidelines and specified inclusion/exclusion criteria, by a trained team of speech and language therapists and psychologists. The remaining publications were then quality assessed using the PEDro (Physiotherapy Evidence Database) and Single Case Experimental Design (SCED) critical appraisal tools. Studies were then categorised according to an intervention typology developed from focus group discussion with 40 speech and language therapists.
Two hundred and four publications met the search criteria. Search criteria for study design were broad and included designs such as; ABA, Multiple Baseline, Cohort, Between Groups, and Randomised Control Trial. Two critical appraisal tools were used to allow for differences in study methodology, but both used criteria such as; eligibility criteria being specified, random allocation to groups, allocation concealment, blinding, baseline comparisons, intention to treat analysis, appropriateness of statistical comparisons.
Data from the focus groups with speech and language therapists was analysed thematically, generating ten therapy categories which were used to group the interventions. Data was then extracted, following a realist synthesis model to determine which children in which contexts had participated in which interventions with what outcomes. A total of 17,661 children participated in the identified studies.
The review found a considerable variation in the quality of the research. Recommendations regarding the design and reporting of intervention studies. A systematic map of the evidence will be presented using the ten therapy categories as the basis for the analysis. This provides graphic evidence of the strength of evidence for each intervention category and the gaps in the evidence base underpinning practice. The findings of the review will be discussed in light of their implications for care pathway design.